Forum OpenACS Development: Edit this Page Release 1.0 - http://etp.museatech.net

Hey everybody,

I'm announcing the release of Edit this Page (ETP) 1.0, a new CMS for OpenACS 4 built by Luke Pond with important help from Dave Hill and Pat Colgan. To get the APM, read the documentation and see some demos, visit http://ETP.museatech.net

Luke's description of ETP can be found at the site, but here it is for those eager to read it before they open their browsers:

Background

I tried using the ArsDigita CMS that was included in ACS 4.2 and 
found myself wanting something a lot more stripped down and easier to 
learn. I spent about 8 hours on it and never really understood how to 
get it to actually deliver a page with some content that I'd entered. 
It frustrated me that in order to create content types of my own, I'd 
have to extend the data model with new tables, views, and pl/sql 
routines. I also thought that it was poorly integrated with the rest 
of the ACS: it stored its own folder hierarchy rather than using the 
Site Map feature like every other ACS package. And, yes, the user 
interface with frames and javascript didn't brighten my day. 

In building Edit This Page with OpenACS 4 I tried to overcome these 
problems. It's quick to get started editing content on your website. 
You can add a new content type (extending the list of content 
attributes available to your page template) by calling a single tcl 
function at server startup. And each level of the hierarchy on the 
site you build is another package instance, easily comprehensible by 
looking at the Site Map, in the context of all the other non-CMS 
packages you may be using. 

ETP uses the same data model that evolved along with the ArsDigita 
CMS, which is found in the acs-content-repository package. Installing 
ETP does not create any tables. The "right thing to do" with all 
packages built for OpenACS 4 is to make use of the standard content 
repository tables, and ETP is no exception. It therefore benefits 
from the attention paid by the community to this data model, which 
includes search engine integration among other good things. 
Currently, ETP only works with OACS4 for PG.

I hope everyone in the community gives it a run! As with everything in OACS 4, it needs some testing and suggestions for improvements. So we're eager to hear your comments!

talli

The first thing I would like to see is a way to specify templates for a section or page from the web UI.  I didn't see a way to do this. Or is there a good reason it is not in there?
Hi Dave, the web UI is primarily intended for use by those whose job it is to edit the content on the site.  So it presents very little in the way of configuration options, in hopes of making the UI more approachable.  The programmer who sets up the site initially uses some tcl procedures to extend the list of "ETP applications" that determine the templates and content types used within each ETP package instance.
Is this the place to ask questions? I am having a problem in the new application. A query is comparing sysdate() which according to postgresql does not exist as a function. I installed from a CVS checkout of OpenACS 4 from yesterday (9/1/01).

The news application works at etp.museatech.net. I can't find sysdate being defined as a function anywhere in the postgresql datamodel.

This is so...cosy!

One improvement I'd like to see though is the ability to browse and upload files from the filesystem as a complement to entering content through the TEXTAREA there is today.

Where is all the content stored? In the content repository? Can you choose to store it in the filesystem? Can you combine filesystem storage and db storage at the same time under the one and same content section?

Will ETP be included in the future distributions of OpenACS 4 and hosted in "our" SDM/CVS-tree?

I'm sure I will have even more questions and suggestions soon enough as I begin to take advantage of this promising system...

sysdate() is defined in www/doc/sql/postgres.sql with OpenACS 3.x, along with things like the 'dual' view.  Perhaps they were running on a DB upgraded from 3.x...